OKCupid

Mar. 25th, 2004 08:40 pm
danaeris: (Default)
[personal profile] danaeris
Damn this shit is addictive.

Why people use OKCupid

I think a lot of people join not because they expect to meet anyone or ever actually date someone they meet on there, but because it is a fun silly thing to do. I've really been enjoying answering the questions and seeing who I know is rated where for me.

I also have to admit that the few people who have really consistently wound up in my top ten who I don't know intrigue me. Not in a hummina hummina way, but more in a, "I think I should know you and I don't!" kinda way.

Two methods of using the system

OKCupid uses these questions to match you up with people. You give your own answer, and then you select the answers you find acceptable, and then select how important this thing is, from irrelevant to mandatory.

So there are two methods, one of which will probably eliminate some good matches but also is more likely to find your "Ideal Match." The other method is more likely to give you a pool of people that meet minimum requirements.

Basically, whenever there is a question that has several acceptable answers, rather than picking the one you'd prefer your match to have, you can select all the acceptable answers and rate it as mandatory. This eliminates the losers who choose the one you don't like, but includes the mediocre ones who selected the mediocre answers.

I've mostly been doing that method. I think it would be interesting for someone who had the time to start two accounts and answer questions one way (only voting for the ideal answer) in the first account, and the other way (voting for all acceptable answers) in the second account. Then we could see how the matches differed, and reach some conclusions about which strategy is more effective towards which goals.

Ratings:
91% [livejournal.com profile] joedecker
89% underwatercolor, [livejournal.com profile] rosefox, [livejournal.com profile] sinboy
88% [livejournal.com profile] dragon_spirit, [livejournal.com profile] elfwreck, lazuli
87% [livejournal.com profile] gerardp, [livejournal.com profile] darkmoon, calico_j, [livejournal.com profile] uncledark, powerofslack, [livejournal.com profile] ianh, [livejournal.com profile] mactavish, [livejournal.com profile] vvvexation, [livejournal.com profile] basmati
(Now only listing people I know)
86% [livejournal.com profile] weirdodragoncat, [livejournal.com profile] luwenth
85% [livejournal.com profile] deyo, [livejournal.com profile] atziluth, [livejournal.com profile] _duncan,

Date: 2004-03-25 09:16 pm (UTC)
From: [identity profile] unseelie23.livejournal.com
The other thing, of course, is that if you only heavily weight the poly related questions, you will skew heavily towards poly, but likely otherwise incompatible. It doesn't hurt to pick a few other questions that are non-poly, non-sex and make those more important.

Date: 2004-03-25 09:27 pm (UTC)
From: [identity profile] phantomdancer.livejournal.com
Heh, I don't have a single match as high as all the ones you just posted...

Date: 2004-03-26 01:54 pm (UTC)
From: [identity profile] truthspeaker.livejournal.com
I was reading their FAQ, and I'm impressed with their approach and algorithms. For example, a mandatory question isn't an absolute filter, but worth 250 times something you care little about. Of course, my first reaction was "That's not mandatory!" but after I read what they had to say about it and thought about it for a while, it made sense to me. Minor errors or slip-ups occur and should get lost in the wash rather than eliminating you because you answered 1 out of 1000 questions incorrectly, possibly by mistake. However, it takes a lot to overcome 1 mandatory question -- 250 is rather a lot.

Profile

danaeris: (Default)
danaeris

August 2022

S M T W T F S
 123456
78910111213
14 151617181920
21222324252627
28293031   

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jan. 22nd, 2026 12:32 am
Powered by Dreamwidth Studios